laser doppler velocimetry · 2016. 6. 29. · • laser light reflected/scattered by moving blood...
TRANSCRIPT
-
BioMedical Optics
FdM 1
Biomedical Optics
Laser Doppler Velocimetry
F.F.M. de Mul
(University of Twente, Enschede, the Netherlands)
-
BioMedical Optics Biomedical Optics : course
FdM 2
Contents
1. - General Introduction
- Overview of existing techniques
2. - Light scattering, theoretical background
- Monte-Carlo + numerical assignment
- Photoacoustics
3. Experimental: focus on some techniques:
- Laser-Doppler perfusion
- Self-mixing velocimetry
-
BioMedical Optics Laser Doppler Velocimetry
FdM 3
Contents:
1. LD for blood perfusion • Principles
• Monitoring
• Imaging
2. Self-mixing LD • Principles
• Experimental aspects
• Flow velocities
• Intra-arterial use
-
BioMedical Optics LDV: Principles
FdM 4
Principle of laser Doppler:
vkk 0s
D
k0 and ks : incoming and
scattered wavevectors
v: particle velocity
D = 2 fD: Doppler frequency k0
ks
v k = ks-k0
21sin.2kk
cossin21
kvfD
Normally in tissue: is small : 0.95 < 15º
approx. k0 // v : only v-component k0 measured
-
BioMedical Optics LDV: Types of Instruments (1)
FdM 5
(A) Differential (Dual-beam) Laser Doppler Velocimetry
Interference fringe pattern
Detection preferentially in
forward direction Scattered
intensity
burst
time
sin2 spacing Fringe s
/sin.2 frequencyDoppler
zvf
z
-
BioMedical Optics LDV: Types of Instruments (2)
FdM 6
(A) Differential (Dual-beam) Laser Doppler Velocimetry
Original frequency
[ 1014 Hz]
Doppler-shifted
frequency
+ D [ 1014 Hz]
Doppler frequency
D [
-
BioMedical Optics LDV: Types of Instruments (3)
FdM 7
(B) Laser Doppler Perfusion Velocimetry
laser
detector v
0
0
Averaged Doppler frequency:
Heterodyne mixing
= - 0
laser
detector v
0
1 2 Homodyne mixing
= 1 - 2
Averaging by: - random velocities
- (multiple) scattering in random directions
-
BioMedical Optics LDPV: Spectra (1)
FdM 8
(B) Laser Doppler Perfusion Velocimetry
0 D
f(D)
Doppler frequency spectrum
Heterodyne
peak
Heterodyne peak: due to large amount of non-Doppler shifted
scattered photons.
Doppler power spectrum
0 D
homodyne
heterodyne S(D)
-
BioMedical Optics LDPV: Spectra (2)
FdM 9
(B) Laser Doppler Perfusion Velocimetry
Doppler power spectrum
0 D
S(D)
Larger v
Larger c
Power spectrum
dependent on:
- concentration c
- velocity v
0
)(.
:spectrumPower
of Moments
dSM nn
-
BioMedical Optics LDPV: Spectra (3)
FdM 10
(B) Laser Doppler Perfusion Velocimetry
Doppler power spectrum
0 D
S(D)
Power spectrum dependent on:
- concentration c
- velocity v
0
)(.
:spectrumPower of Moments
dSM nn
Moments: n = 0 : ~ concentration of moving scatterers
n = 1 : ~ flux of moving scatterers
M1 / M0 : ~ velocity
-
BioMedical Optics LDPV
FdM 11
(B) Laser Doppler Perfusion Velocimetry
Laser
Photodiode
fiber
-
BioMedical Optics LDPV: Skin Tissue
FdM 12
(B) Laser Doppler Perfusion
Velocimetry
Schematic cross section of
Skin tissue
-
BioMedical Optics LDPV: Signals
FdM 13
(B) Laser Doppler Perfusion Velocimetry
LD spectra of
finger tip
upon occlusion
of upper arm
Flux = 1st moment of power spectrum
Concentration = 0th moment
= flux/concentration
heart beat
heart beat
noise
20 sec
occlusion
-
BioMedical Optics LDPV: Types of Instruments
FdM 14
(B) Laser Doppler Perfusion Velocimetry: Instrument design
laser detector
v
0
glass fibers (a) Glass fibers for light
transport
v
probe (b) Direct-contact velocimetry:
Laser and detector in one probe
Disadvantages:
• motional artefacts
• sensitive for local variations
• no motional artefacts
• local averaging
-
BioMedical Optics LDPV: New Instruments
FdM 15
(B) Laser Doppler Perfusion Velocimetry: Instrument design
LD-monitor on-a-chip
provides miniature depth-sensitive
sensor.
Green: photodiode rows
Blue/red: electronics:
amplifiers/multiplexers
Red dot in yellow area:
VCSEL- laser diode
-
BioMedical Optics Clinical applications: (1) Plastic Surgery
FdM 16
(B) Plastic and Reconstructive Surgery
Replantation and
Revascularization
N=68
Flaps: forearm, toe,
thumb, lateral arm,
fibula
Effects after
first half hour
Courtesy: L. van Adrichem, University Hospital Rotterdam
Resulting Inter-
vention level
(“alarm value”)
= 10
-
BioMedical Optics Clinical applications: (1) Plastic Surgery
FdM 17
(B) Plastic and Reconstructive Surgery
Courtesy: L. van Adrichem, University Hospital Rotterdam
Specificity and
Sensitivity of
LD Perfusion flowmetry
in the
replantation/
revascularization
group
“alarm value”,
intervention
level
-
BioMedical Optics Clinical applications: (1) Plastic Surgery
FdM 18
(B) Plastic and Reconstructive Surgery
Characteristics of
Post-operative
monitoring devices
Courtesy: L. van Adrichem, University Hospital Rotterdam
Figures:
instruments read-out
Red line:
Intervention level
-
BioMedical Optics Clinical applications: (1) Plastic Surgery
FdM 19
(B) Plastic and Reconstructive Surgery
Courtesy: L. van Adrichem, University Hospital Rotterdam
Cigarette smoking:
1.
Effect on flow through
healthy thumb.
2.
Effect on flow through
replanted digit.
-
BioMedical Optics Clinical Applications (2): PORH: Post-Occlusive Reactive Hyperemia
0
50
100
150
200
250
300
350
0 500 1000 1500 2000
time (sec)
perf
usio
n
FdM 20
Resting flux
PORH occlusion
Healthy person
-
BioMedical Optics PORH:
2002-01-25_Reindert: Flux in Dorsum with Perimed0.25
-10
0
10
20
30
40
50
60
0 200 400 600 800 1000 1200 1400 1600 1800 2000
2001-08-10_Meijer: Flux in Dorsum with Perimed0.25
-5
0
5
10
15
20
0 200 400 600 800 1000 1200 1400 1600 1800 2000
2001-08-02_Smit: Flux in Dorsum with Perimed0.25
-10
0
10
20
30
40
50
60
70
80
90
0 200 400 600 800 1000 1200 1400 1600 1800 2000
FdM 21
Diabetes Mellitus:
Medium rise, low top
Control group:
Fast rise, high top
PAOD:
Slow rise, low top
-
BioMedical Optics
PORH: Post-Occlusive Reactive Hyperemia
FdM 22
Occlusion
position
Measurement
position
Leg arteries and veins
Capillaries
and shunts
Leg arteries and veins:
• Resistance R1
• Capacitance or
compliance C1
Capillaries/ shunts:
• Resistance R2 .. R4 • Capacitance or
compliance C2
Model: jump-response
after occlusion:
R1 R2
R3 R4 C1 C2 V1 V0 V2
-
BioMedical Optics
PORH: Post-Occlusive Reactive Hyperemia
hE
lrC
r
lR
2
3;
83
0
4
0
hEr
lRC
0
212
FdM 23
Model: jump-response after occlusion:
R1 R2
R3 R4 C1 C2 V1 V0 V2
V = pressure (voltage) [N/m2]
I = flow (current) [m3/s]
Q = volume (charge) [m3]
R = resistance = V / I [Ns/m5]
C = capacitance = Q / V [m5/N]
l = length of tube r0 = radius
= viscosity [Ns/m2] h = wall thickness
E = Young’s modulus [N/m2]
RC- circuits behave in time like : exp(-t / ): with characteristic time constant .
-
BioMedical Optics Post-Occlusive Reactive Hyperemia
214121
12
/
2
/
1
..,..,with
1;..1 21
CCRRfkk
kkekekIItt
rest
21 // 111 ttrestapprox eMReII FdM 24
Model: jump-response after occlusion:
R1 R2
R3 R4 C1 C2 V1 V0 V2
Exact model:
Measured perfusion:
Current I through R2.
Approximation: R4 >> R1 .. R3 (no shunts before entrance in capillary bed)
R1 C1
-
BioMedical Optics Post-Occlusive Reactive Hyperemia
2
32
32
3222111 ;;
R
RRMR
RR
RRCCR
21 // 111 ttrestapprox eMReII
FdM 25
Model: jump-response after occlusion:
R1 R2
R3 R4 C1 C2 V1 V0 V2
Measured perfusion:
Current I through R2.
Approximation: R4 >> R1 .. R3 (no shunts before entrance in capillary bed)
R1 C1
-
BioMedical Optics Post-Occlusive Reactive Hyperemia
21 // 111 ttrestapprox eMReII
FdM 26
Model: jump-response after occlusion:
Approximation:
Measured perfusion:
Current I through R2.
t
1
0
MR
e -t/τ1
1-e-t/τ1
Assume: τ1
-
BioMedical Optics Post-Occlusive Reactive Hyperemia
FdM 27
Exact model Approximation
-
BioMedical Optics Post-Occlusive Reactive Hyperemia
tR tM tH M/R
PAOD 6.5 58 149 3.1
Controls
-
BioMedical Optics Clinical applications (3): Iontophoresis
FdM 29
tissue
Probe with
channel for
drug delivery
Measured:
Influence of drug delivery on flow.
Amp-meter measures amount of injected molecules.
Measure for diffusion coefficient of drug in tissue.
Fibers to LDF instrument
electrode
A Ampère-meter
-
BioMedical Optics Clinical applications (3): Iontophoresis
FdM 30
Measured:
Influence of drug delivery on flow.
Amp-meter measures amount of injected molecules.
Measure for diffusion coefficient of drug in tissue.
Example:
Women with pre-eclampsia (pregnancy poisoning):
- hypertension, proteinuria, oedema,
- due to endothelial dysfunction, changes in vascular reactivity and
permeability for macromolecules
Vasodilatation can be enforced by drugs:
- endothelial-dependent drugs: acetylcholine
- endothelial-independent drugs: nitroprusside
Question: relation between flow change and drug delivery.
-
BioMedical Optics Clinical applications (3): Iontophoresis
FdM 31
Sponge pad
LDF-probe
(reference)
LDF-probe
(measurement)
-
BioMedical Optics
Clinical applications (3): Iontophoresis
Compounds in iontophoresis Activity
Capsaicin Neuropeptides
(Nor)epinephrine hydrochloride Vasoconstrictor
Sodium nitroprusside Vasodilator, to smooth muscle walls
Acetylcholin chloride Vasodilator, activating endothelial
vessel cells
Histamine Oedema and vasodilation
FdM 32
-
BioMedical Optics Clinical applications (3): Iontoforesis
FdM 33
-20
0
20
40
60
80
100
120
1 181 361 541 721 901 1081 1261 1441 1621 1801 1981
-50
0
50
100
150
200
250
300
1 212 423 634 845 1056 1267 1478 1689 1900 2111 2322
Per
fusi
on
Time (sec) Time (sec)
Acetylcholin chloride Sodium nitroprusside
5 Shots: Start ( ↓ ) at 600 sec, duration 20 sec each, intervals 90 sec
End ( ↑ ) at 960 sec
At end of recording: arterial occlusion.
Shown signal = measuring probe – reference probe
Acetylcholine washes out faster, due to vasodilation,
especially with women with pre-eclampsia.
-
BioMedical OpticsLD - Diffusion model for Iontophoresis 1. 1-Dimensional diffusion : c = concentration
2
2
z
cD
t
c
,
}4
exp{),(2
Dt
z
Dt
Qtzc
2. Decrease too slow: add decay term: - c
cz
cD
t
c
2
2
}4
exp{),(2
tDt
z
Dt
Qtzc
Assume:
excess (*) flow
F(t) ~ concentration (*) = above resting
flux level.
tntt
tDt
z
Dt
QsntF
n
n
nn
N
n
)1(
4exp)1(exp)(
20
1
3. Assume: N shots with interval t
s = shot saturation constant;
if s = 0: all shots contribute equally
if s >>1: only first shot contributes
D = diffusion
constant
= decay constant
-
BioMedical OpticsLD - Diffusion model for Iontophoresis
0
50
100
150
200
250
300
350
400
450
500
0 500 1000 1500 2000 2500
time (sec)
pe
rfu
sio
n (
a.u
.)
,
Fit results (2 Pre-eclampsia patients; SNP-administration): 9 shots; 90 sec. apart.
Parameters: χ2red = 1.15; 1 = 76 ± 10 s ; 2 = 408 ± 30 s ; s = 0.029 ± 0.014
χ2red = 1.90; 1 = 65 ± 12 s ; 2 = 252 ± 30 s ; s = 0.048 ± 0.028
-
BioMedical Optics SM-LDV: Self-mixing (1)
FdM 36
Principle:
• laser light reflected/scattered
by moving blood cells,
• partly back-reflected into laser cavity,
• with Doppler-shifted frequency,
• in cavity: mixing with “original” light,
• Doppler signal results,
• can be measured with photodiode
(C) Self-mixing Laser Doppler Velocimetry: Principle
Laser diode
Optical fiber
Photodiode
Moving cell
Velocity
-
BioMedical Optics SM-LDV: Self-mixing (2)
FdM 37
(C) Self-mixing Laser Doppler Velocimetry: Principle
Five-mirror setup:
- M1 and M2 : facets of laser crystal
- M3 and M4 : facets of fiber
- M5 : reflection at / scattering in moving object MO
MO: moving object
L1 and L2 : lenses
DL: diode laser +
photodiode
M1 M2 M3 M4 M5
fiber L1 DL L2
MO
-
BioMedical Optics SM-LDV: Self-mixing (3)
FdM 38
(C) Self-mixing Laser Doppler Velocimetry: Principle
0 10 20 30 40 50
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
am
plitu
de
, [m
V]
time, [s]
0 20 40 60 80 100
-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
R rotating wheel
NR non rotating wheel
frequen
cy s
pectr
um
, [m
V]
frequency, [kHz]
Time signal Frequency signal
R
NR
-
BioMedical Optics SM-LDV: Self-mixing (4)
FdM 39
(C) Self-mixing Laser Doppler Velocimetry:
Filter for directly reflected light
Reflected light will not be focussed onto laser crystal facet.
Only light returning through the fiber will be fed back into the laser cavity.
Laser
facet
No filter
-
BioMedical Optics SM-LDV: Self-mixing (5)
FdM 40
(C) Self-mixing Laser Doppler Velocimetry
0 100 200 300 400 500 600 700 800 900 100020
30
40
50
60
70
80
90
Cut-off frequency
Zero velocity
Doppler signal
Am
plitu
de
(
V)
Frequency (kHz)
Frequency / kHz
Sp
ectr
al i
nte
nsi
ty
Liquid flow: Cut-off frequency
provides maximum flow velocity
-6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6
0.0
0.2
0.4
0.6
0.8
1.0
u/U
max
Fiber position, r (mm)
Flow profile in whole milk,
normalised on maximum value
-
BioMedical Optics SM-LDV: Self-mixing (6)
FdM 41
(C) Self-mixing Laser Doppler Velocimetry
0 200000 400000 600000 800000 1000000
1E-5
1E-4
1E-3
0,01
0,1
frequency s
pectr
um
, a.u
.
frequency, [Hz]
Iliac
artery
Optical
fiber
Catheter
Basket
Branching in iliac artery of healthy pig
Cut-off frequency at 400 kHz corresponds with a velocity of 16 cm/s.
(Independent measurement using an electromagnetic probe: 14.5 1.0 cm/s)(
Flow – no flow
Cut-off frequency
-
BioMedical OpticsLaser Doppler Perfusion: Monitoring vs. Imaging
FdM 42
laser
detector
monitoring
fibers laser
detector
Scanning mirror
imaging
Scattering at moving cells causes Doppler frequency shift
-
BioMedical Optics
Laser Doppler Perfusion (Imaging)
FdM 43
Superficial perfusion of the
dorsal side of the hand,
characters UT written using
muscular balm.
Upper left: perfusion, not
normalized;
Upper right: DC-reflection
from tissue;
Lower left: perfusion,
normalized with DC
Lower right: perfusion,
normalized with DC2 .
-
BioMedical Optics
Laser Doppler Perfusion (Imaging)
FdM 44
From: Bornmyr, “Laser Doppler flowmetry and imaging - methodological studies.
Dep of clinical hysiology”,thesis, Malmö, Sweden (1998);
Figure: courtesy: prof. G. Nilsson, Lisca, Linkoping, Sweden)
Typically the highest perfusion is in the boundary around
the ulcer, in inflammatory skin and in granulating tissue
inside the ulcer area.
Perfusion Image of a foot ulcer
-
BioMedical Optics
Laser Doppler Perfusion (Imaging)
FdM 45
The effect of micro-trauma
Insertion of a micro-
dialysis fibre into the
skin.
The dialysis fibre
probe tip causes
hyperperfusion
No hyperperfusion
at the point of
introduction because
the skin is anesthetized.
After 30 minutes the
hyperperfusion is reduced.
(Courtesy: Lisca Sweden)
-
BioMedical Optics
Laser Doppler Perfusion (Imaging)
FdM 46
(Courtesy: Lisca Sweden)
Before treatment After treatment
Immediately One week 8.5 months later
Neo-vascularisation
in tumour area.
Inflammatory
response. Inflammatory response
with excessive perfusion. Back to normal.
Basal cell carcinoma
-
BioMedical Optics
Laser Doppler Perfusion (Imaging)
FdM 47
Flow-Maps of a
Healing Wound
Day 1: wound creation
Day 4
Day 7
Day 10
Day 13
Black inc
marker on skin
Crust formation
in centre of wound Crust off Perfusion returning
to normal (Courtesy: Lisca Sweden)
-
BioMedical Optics
Laser Doppler Perfusion (Imaging)
FdM 48
Day 2
Reduced perfusion
in burnt areas.
Increased perfusion
in surrounding skin.
Towards normalisation.
Day 13 Day 28
The healing process of a burn wound
(Courtesy: Lisca Sweden)
-
BioMedical Optics Laser Doppler Perfusion (Monitoring) New developments
FdM 49
Low-coherent depth-sensitive LD-Monitoring
-5 0 5 10 15
0.0
0.2
0.4
0.6
0.8
1.0
experimental
MC simulation
No
rma
lize
d in
tensity
Optical path length [mm]
50:50
95:5
D
SLD
R
The reference mirror selects the depth in the sample from which a coherent Doppler-
shift signal will be measured.
-
BioMedical Optics
Laser Doppler Perfusion (Imaging) New developments:
FdM 50
Imaging using CMOS camera
to PC
CMOS
camera
lens
scattered light
sample
HeNe laser
illuminating beam
Remarks: - Beam diameter on the sample.....5 cm
- Sample-to-camera distance..........60cm
Advantage of CMOS over CCD camera:
CCD: serial read-out of collected photons
using shift registers
=> slow read-out
=> no dynamic response possible
CMOS: each pixel can be addressed separately,
enables fast dynamic 2D-response
Advantage of CMOS over scanning detector:
Scanning detector:
= measures dynamic signal
= slow imaging due to scanning
-
BioMedical Optics
Laser Doppler Perfusion (Imaging)
FdM 51
New development:
Imaging using CMOS camera
Sample : Plastic box with 4 cylindrical holes
filled with Intralipid™ (moving particles)
35
mm
Full frame
Moving particles concentration, a.u.
0 255
Original False-color Smoothed
-
BioMedical Optics
Laser Doppler Perfusion (Imaging)
FdM 52
New development:
Imaging using CMOS camera
Occlusion
After
image processing
Before
image processing
No occlusion
Moving blood cells concentration, a.u.
0 255
-
BioMedical Optics Laser Doppler Velocimetry
FdM 53
The End
Summary:
1. LD for blood perfusion • Principles
• Monitoring
• Imaging
2. Self-mixing LD • Principles
• Experimental aspects
• Flow velocities
• Intra-arterial use